GTM Sauce

The Superwall Podcast · 2025-12-09

Julian Alvarez - This ai study app makes $100k/mo. here's how (no-code)

Joseph Choi interviews Julian Alvarez, co-founder of Jungle, an AI study app at $80-100K/month. Extracted: first users from replying to a viral tweet and DMing everyone who engaged, AI tool directories, micro-influencer deals (one creator took MRR from $2K to ~$15K, but results did not repeat), a UGC program of ~30 creators posting ~400 videos/week at ~$2 CPM, upload-from-landing-page onboarding, a growing-tree mechanic (+70% engagement) and 30-40% of new users from word of mouth.

Play the episode

12 tactics

Jungle · B2C · AI study app

200+ downloads and 50-100 DMs; confirmed the idea
“That tweet blew up, and I commented saying we were building like something similar, which we were building something tangentially similar, but not the exact same thing.”
Workedorganic-socialvalidation
The Superwall Podcast · 2025-12-09

Jungle · B2C · AI study app

First 100 users and lots of feedback; Twitter account got banned
“we basically got our first 100 users by DMing everyone that engaged with that tweet, so much to the point that my Twitter account got banned, which was really sad.”
Mixedoutbound-salesfeature-development
The Superwall Podcast · 2025-12-09

Jungle · B2C · AI study app

Creators found it and made videos for free; guest thinks directories are now saturated
“posting on the AI tool directories and having it get picked up by some influencers, there were some videos that started to go a little bit viral and that got a lot of people into the door”
Workedpr-launchinfluencer
The Superwall Podcast · 2025-12-09

Jungle · B2C · AI study app

Led to the directory and creator traction; founders say it changed their mindset
“Now, for the next 2 weeks, don't touch the product. Just spend the next 2 weeks focusing purely on distribution, and nothing happens, then no one probably cares about your product and you should probably pivot.”
Workedvalidationteam-process
The Superwall Podcast · 2025-12-09

Jungle · B2C · AI study app

One creator took MRR from $2K to ~$15K; couldn't repeat it and lost money
“target micro to mid-tier influencers, like in the 5k to 100k followers range, because the ROI and the CPMs at that time I think are a lot better versus like if you get really big influencers”
Mixedinfluencercreator-ops
The Superwall Podcast · 2025-12-09

Jungle · B2C · AI study app

~400 videos a week at ~$2 CPM; seems profitable, but the tactic is saturating
“They create their own separate accounts, and we basically have them post like 10 to 12 videos a week. Very high content throughput. At like 30 creators, that means we're getting like close to 400 videos per week.”
Workedugc-creatorscreator-ops
The Superwall Podcast · 2025-12-09

Jungle · B2C · AI study app

One of the most successful of hundreds of landing page tests; no numbers
“we've iterated and run hundreds of experiments on our landing page that you're showing here, and one of the most successful ones is basically being able to upload a document or a URL directly from the landing page.”
Workedlanding-pageproduct-activation
The Superwall Podcast · 2025-12-09

Jungle · B2C · AI study app

Guest says fast time to value drives word of mouth; no separate numbers
“Like we don't show you all these options of like what range of minutes or pages do you want to generate for, what type of questions do you want, how many questions do you want.”
Workedonboardingproduct-activation
The Superwall Podcast · 2025-12-09

Jungle · B2C · AI study app

Core engagement up 70%; onlookers in libraries ask about it
“One, that tree increased engagement by 70%, basically, like the core engagement metric of answering a lot of questions. But it's also a really good growth hack.”
Workedretentionproduct-virality
The Superwall Podcast · 2025-12-09
“if you show a paywall, that means that 50% of people that complete the mini review or do this first experience will see the paywall, versus if you only show the paywall after the first time you generate and only like 25% of people get there”
Unknownpaywallretention
The Superwall Podcast · 2025-12-09
“So one thing we also strategically do is like we split our onboarding into three main phases.”
Unknownonboardingnotifications
The Superwall Podcast · 2025-12-09

Jungle · B2C · AI study app

30-40% of new users come from word of mouth
“It's like product, right? Building a really good product, trying to improve retention as much as we can, increasing shareability, and unlocking viral growth loops within the app.”
Workedword-of-mouthproduct-virality
The Superwall Podcast · 2025-12-09

Numbers mentioned

SubjectMetricValueQuote
JunglerevenueUSD 80K–100K/month
approximate
“We are making around like 80 to 100k a month.”
Jungledownloads
Over 200, from commenting on a viral tweet
200 downloads
approximate
“I got like over 200 downloads, over 50 to 100 DMs”
Jungleactivity volume
Inbound DMs after the tweet reply
50–100 DMs received
approximate
“I got like over 200 downloads, over 50 to 100 DMs”
Jungleusers
First 100 users, from DMing everyone who engaged with the tweet
100 users
approximate
“we basically got our first 100 users by DMing everyone that engaged with that tweet, so much to the point that my Twitter account got banned”
Jungleusers
10,000+ users, mainly from influencer marketing
10K users
approximate
“getting to 10,000+ ended up coming from influencer marketing”
Junglefollowers
Target follower range for micro to mid-tier influencers
5,000–100K followersper creator“target micro to mid-tier influencers, like in the 5k to 100k followers range”
personAgogo, medical-student creatorviews
About three of her Jungle videos passed 1M views each
1M viewsper video
approximate
“she had maybe like three videos that got over a million views, this being one of them”
Junglemrr
From ~$2K to almost $15K MRR within 1-2 weeks of Agogo's first viral video
USD 15K
approximate
“we went from 2k MRR to like almost 15k MRR within like a week or 2”
Junglerevenue
Revenue attributed to Agogo's single viral video
USD 20Kper video
estimate
“that video alone, I think we estimate made us like close to 20k or so”
Jungleroas
Influencer marketing after the first viral creator
below 1: spent significantly more than got back
estimate
“Yeah, definitely seemed like we were spending significantly more money than we were getting back.”
Jungleshare
Share of new users from word of mouth, measured by a 'How'd you hear about us?' onboarding question
30–40%
approximate
“Yeah, around like 30 to 40% of all new users come from word of mouth.”
Junglecount
Experiments run on the landing page
hundreds
approximate
“we've iterated and run hundreds of experiments on our landing page”
Jungleengagement rate
Lift in the core engagement metric (questions answered) from the growing-tree visual
+70%
approximate
“One, that tree increased engagement by 70%, basically, like the core engagement metric of answering a lot of questions.”
benchmarkMinimum weekly new users an early app needs to learn from experimentsnew customers
At minimum; ideally many more
30–100 users/week
approximate
“I think early on, you got to figure out a way to consistently get at least 30 to 100 people in the door per week”
Jungleactivity volume
Posting quota for each UGC creator account
10–12 videosper creator /week
approximate
“They create their own separate accounts, and we basically have them post like 10 to 12 videos a week.”
Junglecount
UGC creators in the program
30 creators
approximate
“At like 30 creators, that means we're getting like close to 400 videos per week.”
Jungleactivity volume
Total UGC videos per week (close to 400)
400 videos/week
approximate
“At like 30 creators, that means we're getting like close to 400 videos per week.”
Junglecount
Description/teaser figure ('40 UGC creators posting 400 videos per week'); the guest says about 30 in the body
40 creators
approximate
“running 40 UGC creators posting 400 videos per week”
Junglecpm
UGC program cost per 1,000 views; varies a bit; much better than influencer marketing
USD 2per 1000 views
approximate
“we're able to get basically like around $2 CPMs or so, that varies a little bit up and down, but it's pretty good and way better than influencer marketing.”